Theoretical analysis of a class of randomized regularization methods

Tong Zhang

doi:10.1145/307400.307433

Theoretical analysis of a class of randomized regularization methods

Tong Zhang^*

^*Corresponding author for this work

Research output: Contribution to conference › Conference Paper › peer-review

13 Citations (Scopus)

Abstract

The convergence behavior of traditional learning algorithms can be analyzed in the VC theoretical framework. Recently, many researchers have been interested in a class of randomized learning algorithms including the Gibbs algorithm from statistical mechanics. However, no successful theory concerning the generalization behavior of these randomized learning algorithms have been established previously. In order to fully understand the behavior of these randomized estimators, we shall compare them with regularization schemes for deterministic estimators. Furthermore, we present theoretical analysis for such algorithms which leads to rigorous convergence bounds.

Original language	English
Pages	156-163
Number of pages	8
DOIs	https://doi.org/10.1145/307400.307433
Publication status	Published - 1999
Externally published	Yes
Event	Proceedings of the 1999 12th Annual Conference on Computational Learning Theory (COLT'99) - Santa Cruz, CA, USA Duration: 7 Jul 1999 → 9 Jul 1999

Conference

Conference	Proceedings of the 1999 12th Annual Conference on Computational Learning Theory (COLT'99)
City	Santa Cruz, CA, USA
Period	7/07/99 → 9/07/99

Access to Document

10.1145/307400.307433

Cite this

@conference{e86a56d7d3834e59ad0a692cfe5a6a4b,

title = "Theoretical analysis of a class of randomized regularization methods",

abstract = "The convergence behavior of traditional learning algorithms can be analyzed in the VC theoretical framework. Recently, many researchers have been interested in a class of randomized learning algorithms including the Gibbs algorithm from statistical mechanics. However, no successful theory concerning the generalization behavior of these randomized learning algorithms have been established previously. In order to fully understand the behavior of these randomized estimators, we shall compare them with regularization schemes for deterministic estimators. Furthermore, we present theoretical analysis for such algorithms which leads to rigorous convergence bounds.",

author = "Tong Zhang",

year = "1999",

doi = "10.1145/307400.307433",

language = "English",

pages = "156--163",

note = "Proceedings of the 1999 12th Annual Conference on Computational Learning Theory (COLT'99) ; Conference date: 07-07-1999 Through 09-07-1999",

}

TY - CONF

T1 - Theoretical analysis of a class of randomized regularization methods

AU - Zhang, Tong

PY - 1999

Y1 - 1999

N2 - The convergence behavior of traditional learning algorithms can be analyzed in the VC theoretical framework. Recently, many researchers have been interested in a class of randomized learning algorithms including the Gibbs algorithm from statistical mechanics. However, no successful theory concerning the generalization behavior of these randomized learning algorithms have been established previously. In order to fully understand the behavior of these randomized estimators, we shall compare them with regularization schemes for deterministic estimators. Furthermore, we present theoretical analysis for such algorithms which leads to rigorous convergence bounds.

AB - The convergence behavior of traditional learning algorithms can be analyzed in the VC theoretical framework. Recently, many researchers have been interested in a class of randomized learning algorithms including the Gibbs algorithm from statistical mechanics. However, no successful theory concerning the generalization behavior of these randomized learning algorithms have been established previously. In order to fully understand the behavior of these randomized estimators, we shall compare them with regularization schemes for deterministic estimators. Furthermore, we present theoretical analysis for such algorithms which leads to rigorous convergence bounds.

UR - https://openalex.org/W1988943785

U2 - 10.1145/307400.307433

DO - 10.1145/307400.307433

M3 - Conference Paper

SP - 156

EP - 163

T2 - Proceedings of the 1999 12th Annual Conference on Computational Learning Theory (COLT'99)

Y2 - 7 July 1999 through 9 July 1999

ER -

Theoretical analysis of a class of randomized regularization methods

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this